Both fluoroscopy and computed tomography are becoming increasingly utilized in pediatric medicine. Quantitative methods for determining organ dose in pediatric patients are thus essential for clinical decision-making and risk assessment. Even approximate methods of risk assessment, based upon measures of entrance dose or energy imparted, fundamentally rely upon knowledge or organ dose. Organ doses may be determined through either computational simulation of the diagnostic exam using anthropomorphic models, or through the use of dosimeters embedded within anthropomorphic physical phantoms. Anthropomorphic computational models may be further classified as either stylized models, where organs are delineated by 3D surface equations, or tomographic models, in which organs are determined from segmented CT or MRI images. In this project, the investigators will develop improved techniques for estimating organ doses to the newborn child in both fluoroscopic and CT examinations. The project Specific Aims are: 1) to construct a high-resolution, segmented tomographic computational model of a newborn child using helical CT images of live newborns. The model will be scaled to match the dimensions and organ masses of the MIRD newborn model; 2) to construct a high-resolution tomographic physical phantom of a live newborn using the identical CT data and also scaled to match the MIRD newborn model. A full-scale physical phantom of the stylized MIRD model will also be constructed. Internal organ doses will be assessed using embedded high-sensitivity MOSFET dosimeters; 3) to determine organ doses in the newborn child received during fluoroscopic and CT examinations using both the computational model and physical phantoms developed in Specific Aims 1 and 2, respectively; and 4) to evaluate the degree to which improved anatomic representation, in either computational models or physical phantoms, influences estimates of organ dose in newborn radiological examinations.